Yesterday I did some more work on some CSG-macros of mine.
They will "round off" any concave shape made with intersections
of planes, like the upper left shape.

The lower right shape shows how the resulting shape looks
like if both the edges and corners are wounded off.

The rendering is slow, so now I'm working on the macro
so that it will render much faster.

The upper right shape shows how the rounding looks like if
only the edges are rounded off.

And the lower left shape has different colours where "material"
has been "removed".

When I'm finished with the macros I'll post the source code
if anyone is interested.


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Attachments:
Download 'roundoffplanes22_01_cleaned09.jpg' (43 KB)

Preview of image 'roundoffplanes22_01_cleaned09.jpg'

Tor Olav Kristensen wrote:
> 
> Yesterday I did some more work on some CSG-macros of mine.
> They will "round off" any concave shape made with intersections
> of planes, like the upper left shape.

Nota bene:  The shapes depicted are convex, not concave.

> The lower right shape shows how the resulting shape looks
> like if both the edges and corners are wounded off.
> 
> The rendering is slow, so now I'm working on the macro
> so that it will render much faster.

It's possible that making each face into a prism will help.  This
might be difficult, however.

Regards,
John

Post a reply to this message

"John VanSickle" <van### [at] erolscom> wrote in message
news:39DC9C88.54B9888F@erols.com...
| Tor Olav Kristensen wrote:
| >
| > Yesterday I did some more work on some CSG-macros of mine.
| > They will "round off" any concave shape made with intersections
| > of planes, like the upper left shape.
|
| Nota bene:  The shapes depicted are convex, not concave.

Well, not if you put the camera inside  :-)
You're quite the wizard with this kind of thing Tor, manipulating CSG
objects.  I just wanted to say that.
Any chance this could merge with one of the rock making includes?  Then
again, if this is slow and the rock makers are already slow it might be too
much for the typical scene file.
If the amount of roundedness can be adjusted it certainly could do well for
worn edges of such shapes anyway.

Bob

Post a reply to this message

Tor Olav Kristensen wrote:
> 
> Yesterday I did some more work on some CSG-macros of mine.
> They will "round off" any concave shape made with intersections
> of planes, like the upper left shape.
> 
> The lower right shape shows how the resulting shape looks
> like if both the edges and corners are wounded off.
> 

looks good, how did you do it ? just CSG with planes, cylinders and spheres?

> The rendering is slow, so now I'm working on the macro
> so that it will render much faster.
> 

I suppose parsing will also be quite slow when working with complicated shapes.  

> The upper right shape shows how the rounding looks like if
> only the edges are rounded off.
> 
> And the lower left shape has different colours where "material"
> has been "removed".
> 

> When I'm finished with the macros I'll post the source code
> if anyone is interested.
> 

Sure we are :-)
I estimate it's quite a lot of work doing it, so thanks in advance...

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
Homepage: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

I want the coordinates for a truncated cube, in terms of triangles.

Can anyone point me to a reference for this?

In other words, I have seen the list of triangles needed for a cube/
rectangular solid based on 2 corner points.  I want 2D projection of this
solid to be, instead of a square, a symmetric 8-sided figure.

Post a reply to this message

John VanSickle wrote:
> 
> Tor Olav Kristensen wrote:
> >
> > Yesterday I did some more work on some CSG-macros of mine.
> > They will "round off" any concave shape made with intersections
> > of planes, like the upper left shape.
> 
> Nota bene:  The shapes depicted are convex, not concave.

Ugghh...  -How did that happen ?!

I even looked up convex in the dictionary 
to make sure that this wasn't only a 
norwegian word before I wrote the post.


> > The lower right shape shows how the resulting shape looks
> > like if both the edges and corners are wounded off.
> >
> > The rendering is slow, so now I'm working on the macro
> > so that it will render much faster.
> 
> It's possible that making each face into a prism will help.
> This might be difficult, however.

Yes I think you're right. I like that idea.
Prisms would probably make it faster.

It would not be very difficult. The needed 
data are already available within my macros.
But it will be some work determining how
data should be passed between the different 
macros in an efficient way.

There's one problem with prisms though:
The "stones" could not be used in further
CSG operations. E.g. like I did in my image; 
"Two spheres" posted to this news group 26.
April:

news://news.povray.org/3906D341.F1CD358C%40online.no

The two metal spheres are put inside the 
"space" cut out by a convex shape "rounded off"
by some preliminary versions of these macros.


Tor Olav

-- 
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Bob Hughes wrote:
> ...
> You're quite the wizard with this kind of thing Tor, manipulating CSG
> objects.  I just wanted to say that.

Thank you !  - I'm flattered :)


> Any chance this could merge with one of the rock making includes?  Then
> again, if this is slow and the rock makers are already slow it might be too
> much for the typical scene file.

I'll look into it. Where can I find these 
rock making includes ?


> If the amount of roundedness can be adjusted it certainly could do well for
> worn edges of such shapes anyway.

The roundness can be adjusted by setting 
a "rounding radius" for the corners and
the edges.


Tor Olav

-- 
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Christoph Hormann wrote:
> 
> Tor Olav Kristensen wrote:
> >
> > Yesterday I did some more work on some CSG-macros of mine.
> > They will "round off" any concave shape made with intersections
> > of planes, like the upper left shape.
> >
> > The lower right shape shows how the resulting shape looks
> > like if both the edges and corners are wounded off.
> >
> 
> looks good, how did you do it ? just CSG with planes, cylinders and spheres?

The starting (convex) shape is just 
intersections of all the involved planes.

Then one macro takes info about these planes 
and calculates where every possible triple 
of non parallel planes meets at potential 
corners (points).

Then another macro removes all the corners 
that are "false" (I.e. those that have one 
or more planes "below" themselves.)

Then some other macros receives the data 
about the final "true" corners and 
constructs the shapes (CSG with cylinders 
and spheres) that are used to "subtract"
from the starting shape.


> > The rendering is slow, so now I'm working on the macro
> > so that it will render much faster.
> >
> 
> I suppose parsing will also be quite slow when working with complicated shapes.

Yes, but not unbearable slow (as the pre-
liminary macros was).


> > When I'm finished with the macros I'll post the source code
> > if anyone is interested.
> >
> 
> Sure we are :-)
> I estimate it's quite a lot of work doing it, so thanks in advance...

Yes, there's a lot of info to manage within 
the macro. But I've now manage to make them 
much faster. But there's still some work 
left to make them stop calculating the same 
things over and over.

Maybe I can force myself to post some code 
soon, even if I'm not satisfied with it yet.


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

"Greg M. Johnson" wrote:
> 
> I want the coordinates for a truncated cube, in terms of triangles.
> 
> Can anyone point me to a reference for this?
> 
> In other words, I have seen the list of triangles needed for a cube/
> rectangular solid based on 2 corner points.  I want 2D projection of this
> solid to be, instead of a square, a symmetric 8-sided figure.

I'm not sure if I understand what you mean.

Can you try to explain more or post an 
image of this shape (e.g. made by inter-
section of planes) ?

By truncated cube, do you mean that the 
original corners are "cut" off ?


Tor Olav
-- 
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

"Tor Olav Kristensen" <tor### [at] hotmailcom> wrote in message
news:39E0F73A.B0E029C8@hotmail.com...
|
| I'll look into it. Where can I find these
| rock making includes ?

I'm having no luck finding the rock includes or macros, sorry.  One thing is
John Van Sickles and it seems Chris Huff may have something too (not sure),
and at the POV Objects Collection...  Steve Pidgeon was another person that
did one I know.

Bob

Post a reply to this message

"Greg M. Johnson" wrote:

> I want the coordinates for a truncated cube, in terms of triangles.

If you have a cube going from <-1,-1,-1> to <1,1,1> and you want to truncate
it so each of the original six faces becomes a regular octagon, then the
vertices would be at

<±1,±1,±(sqrt(2)-1)>
<±1,±(sqrt(2)-1),±1>
<±(sqrt(2)-1),±1,±1>

(2^3+2^3+2^3=24 vertices)

--
David Fontaine  <dav### [at] faricynet>  ICQ 55354965
My raytracing gallery:  http://davidf.faricy.net/

Post a reply to this message

"Bob H." wrote:
> 
> "Tor Olav Kristensen" <tor### [at] hotmailcom> wrote in message
> news:39E0F73A.B0E029C8@hotmail.com...
> |
> | I'll look into it. Where can I find these
> | rock making includes ?
> 
> I'm having no luck finding the rock includes or macros, sorry.  One thing is
> John Van Sickles and it seems Chris Huff may have something too (not sure),
> and at the POV Objects Collection...  Steve Pidgeon was another person that
> did one I know.
> 
> Bob

If I'm not mistaken John has removed his rock include from his
website but he may have included it with his subdivision surface
suite - http://users.erols.com/vansickl/sss.htm which is in it's
own right a fun toy to play with.

Steve's site is at:
http://www.iro.umontreal.ca/~pigeon/pub/PovPage/RockGen/RockGen.html

See also:
http://www.rahul.net/endl/mallan/html/rocks.htm
http://www-personal.umich.edu/~mstock/pages/rocktools.html

-- 
Ken Tyler - 1400+ POV-Ray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/

Post a reply to this message

On Sun, 08 Oct 2000 23:14:04 -0700, Ken wrote...
> If I'm not mistaken John has removed his rock include from his
> website but he may have included it with his subdivision surface
> suite - http://users.erols.com/vansickl/sss.htm which is in it's
> own right a fun toy to play with.

I had a look at that URL, and it seems that the Surface Subdivision Suite 
page has moved to /dev/null <grin>, as the URL you posted points to a new 
page, which gives a 404...

John? 
 
Bye for now,
     Jamie.

Post a reply to this message

> Yesterday I did some more work on some CSG-macros of mine.
> They will "round off" any concave shape made with intersections
> of planes, like the upper left shape.

Any chance of you posting the macro code itself...

Rarius

Post a reply to this message

Thanks, but what I meant is this:

Take  perfect cube.  set each edge on a grinding wheel such that the normal of
the two faces on this edge make a 45 degree angle with the normal of the
grinding wheel.  Do it for just a second or two.



David Fontaine wrote:

> "Greg M. Johnson" wrote:
>
> > I want the coordinates for a truncated cube, in terms of triangles.
>
> If you have a cube going from <-1,-1,-1> to <1,1,1> and you want to truncate
> it so each of the original six faces becomes a regular octagon, then the
> vertices would be at
>
> <±1,±1,±(sqrt(2)-1)>
> <±1,±(sqrt(2)-1),±1>
> <±(sqrt(2)-1),±1,±1>
>
> (2^3+2^3+2^3=24 vertices)
>
> --
> David Fontaine  <dav### [at] faricynet>  ICQ 55354965
> My raytracing gallery:  http://davidf.faricy.net/

Post a reply to this message

On Mon, 09 Oct 2000 08:27:41 -0400, Greg M. Johnson wrote:
>Thanks, but what I meant is this:
>
>Take  perfect cube.  set each edge on a grinding wheel such that the normal of
>the two faces on this edge make a 45 degree angle with the normal of the
>grinding wheel.  Do it for just a second or two.

So what you want is a cube with beveled edges.  The size of the bevel is a
factor in finding the corners.

What used to be a square face will still be a square face, but it will be 
moved inward on all sides by the size of the bevel.  What used to be an edge
will now be a rectangular face; connect corresponding edges of the new square
faces to find the twelve new faces.  What used to be a vertex will now be a
triangular face; connect corresponding corners of the new square faces to 
find the eight new faces.  The final object will have 26 faces, with a total 
of 44 triangles.

-- 
Ron Parker   http://www2.fwi.com/~parkerr/traces.html
My opinions.  Mine.  Not anyone else's.
Proudly not helping RIAA and SDMI steal my rights -- 
  http://www.eff.org/Misc/EFF/Newsletters/EFFector/HTML/effect13.08.html

Post a reply to this message

Jamie Davison wrote:

> I had a look at that URL, and it seems that the Surface Subdivision Suite
> page has moved to /dev/null <grin>, as the URL you posted points to a new
> page, which gives a 404...

Hmmm.... I'll see if I can track it down.

-- 
Ken Tyler

Post a reply to this message

Ron Parker wrote:

> ...
> So what you want is a cube with beveled edges.  The size of the bevel is a
> factor in finding the corners.
>
> What used to be a square face will still be a square face, but it will be
> moved inward on all sides by the size of the bevel.  What used to be an edge
> will now be a rectangular face; connect corresponding edges of the new square
> faces to find the twelve new faces.  What used to be a vertex will now be a
> triangular face; connect corresponding corners of the new square faces to
> find the eight new faces.  The final object will have 26 faces, with a total
> of 44 triangles.

Are you sure about all this ?   :)

See enclosed image.


To Greg:
When I've cleaned up the source code (that I believe
solves your problem), I'll post it.


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

Attachments:
Download 'truncatedcube03.jpg' (18 KB)

Preview of image 'truncatedcube03.jpg'

Philip Cowley wrote:

> > Yesterday I did some more work on some CSG-macros of mine.
> > They will "round off" any concave shape made with intersections
> > of planes, like the upper left shape.
>
> Any chance of you posting the macro code itself...

Yes I will. -But it needs some more work first.

Right now I'm too tired to do this work.


Tor Olav
--
mailto:tor### [at] hotmailcom
http://www.crosswinds.net/~tok/tokrays.html

Post a reply to this message

"Greg M. Johnson" wrote:

> Thanks, but what I meant is this:
>
> Take  perfect cube.  set each edge on a grinding wheel such that the normal of
> the two faces on this edge make a 45 degree angle with the normal of the
> grinding wheel.  Do it for just a second or two.

Ah, that's bevelled, not truncated :) (Besides i think my data was wrong, it
should have been (2-sqrt(2)) for perfect octagon)
so you want a perfect octagon for the cross section? hmm...

<±1,±(sqrt(2)-1),±(sqrt(2)-1)>
<±(sqrt(2)-1),±1,±(sqrt(2)-1)>
<±(sqrt(2)-1),±(sqrt(2)-1),±1>
<±.75,±.75,±.75> (?)

24 vertices, 18 faces, 60 triangles

--
David Fontaine  <dav### [at] faricynet>  ICQ 55354965
My raytracing gallery:  http://davidf.faricy.net/

Post a reply to this message

"Ken" <tyl### [at] pacbellnet> wrote in message
news:39E1E972.407515C7@pacbell.net...
|
| Jamie Davison wrote:
|
| > the URL you posted points to a new page, which gives a 404...
|
| Hmmm.... I'll see if I can track it down.

Only the last of those 4 addresses worked for me.  No wonder I couldn't
locate anything, and I had misspelled Steven Pigeons' name (sorry Steve).
Anyhow, not sure any of this goes with what Tor has done since the rocks all
seem to be triangle mesh instead of plane intersections.

Bob

Post a reply to this message

"Greg M. Johnson" wrote:

> Thanks, but what I meant is this:
>
> Take  perfect cube.  set each edge on a grinding wheel such that the normal of
> the two faces on this edge make a 45 degree angle with the normal of the
> grinding wheel.  Do it for just a second or two.
> ...

I've now posted some source code that provides you with the coordinates
for the corners. (Watch the output in the debug window.)

See my recent post "Source code for bevelled box" to povray.text.scene-files:

news://news.povray.org/39E3AFCA.B8FB8283%40online.no


Regards,

Tor Olav

Post a reply to this message

"Bob H." wrote:
> 
> "Ken" <tyl### [at] pacbellnet> wrote in message
> news:39E1E972.407515C7@pacbell.net...
> |
> | Jamie Davison wrote:
> |
> | > the URL you posted points to a new page, which gives a 404...
> |
> | Hmmm.... I'll see if I can track it down.
> 
> Only the last of those 4 addresses worked for me.  No wonder I couldn't
> locate anything, and I had misspelled Steven Pigeons' name (sorry Steve).
> Anyhow, not sure any of this goes with what Tor has done since the rocks all
> seem to be triangle mesh instead of plane intersections.
> 
> Bob

After contacting John the Subdivision Surface Suite is now back online
at - http://users4.50megs.com/enphilistor/sss.htm

Steven's page is still valid as is Mack Stocks page. The only one
that appears to be gone is "Rocks" at -

http://www.rahul.net/endl/mallan/html/rocks.htm

for which there appears to be no replacement URL for.

Three out of four ain't bad :)

-- 
Ken Tyler - 1400+ POV-Ray, Graphics, 3D Rendering, and Raytracing Links:
http://home.pacbell.net/tylereng/index.html http://www.povray.org/links/

Post a reply to this message

Oh my dear friend, I am sorry. I wanted triangles.  Let me figure it out myself.  I
am a newbie at discovering the benefits of triangles.    If I were to use the
intersection of a dozen or more planes, I might as well have used a superellipsoid.

Post a reply to this message

Attachments:
Download 'cells.jpg' (21 KB)

Preview of image 'cells.jpg'

On Mon, 09 Oct 2000 08:27:41 -0400, "Greg M. Johnson"
<gre### [at] my-dejanewscom> blathered:

>Thanks, but what I meant is this:
>
>Take  perfect cube.  set each edge on a grinding wheel such that the normal of
>the two faces on this edge make a 45 degree angle with the normal of the
>grinding wheel.  Do it for just a second or two.
>
>

Let me first state that I've read the rest of this thread (as of this
date) and know my answer is not what you're looking for.

This description (above) leads me to describing how I figure out the
appropriate method for creating a shape.  If I were creating a cube
that I wanted to look like it had the corners knocked off with a
grinder, I tend to create the cube, then use a CSG difference to
remove the parts the grinder took away.  In other words, I try to
create the effect of the manufacturing process.

<inside_voice> Pinot Gris has an odd effect on my thought
processes...</inside_voice>

I'd create the cube, then create my imaginary grinding wheel.  Then
I'd use a CSG difference (or maybe a few thousand...) to remove the
corners.



Defective O-O

http://members.home.com/redundant/Defective/
(yes, I know that's the wrong membername...ran out of space at
http://members.home.com/defective/)

Post a reply to this message